Re: PED_MICRO_VERSION, PED_INTERFACE_AGE, PED_BINARY_AGE in configure.in

[K.G.]

On Thu, 10 Nov 2005 15:30:32 +0100 (MET), Szakacsits Szabolcs <address@hidden> 
wrote:

> 
> On Thu, 10 Nov 2005, Otavio Salvador wrote:
> > 
> > > Hopefuly with the big changes that will go in 1.6.26 this
> > > will be the case (or almost the case at the very least ;)
> > > I'll prepare a tag and then ask for review, then for release.
> > > I think there's no problem in displaying Parted 1.6.25 at the
> > > start of 1.6.25.1 because the code don't change, but if
> > > somebody thinks it's better I could directly add ".1" in
> > > the source (i don't think autoconf stuffs support x.y.z.t).
> > 
> > I think 1.6.25 is OK since the codebase is the same and won't matter
> > for bug reporting.
> 
> Two 1.6.25 releases are highly bad. The only right way is to bump up the
> version, document the change so people can know what's going on and act
> accordingly.
> 
>       Szaka
> 
The problem is that 1.6.25 has a soname greater than that
of 1.6.24, so if we release 1.6.26 with the same as that
of 1.6.24 then 1.6.27 with a one greater than both 1.6.24,
1.6.25 and 1.6.26 I doubt the situation will be better than
what we wanted to do? (a 1.6.25.1 package with soname ==
1.6.24 and a version displayed by parted of 1.6.25 or
1.6.25.1, the code being exactly the same anyway).

Guillaume